product-owner
Pass
Audited by Gen Agent Trust Hub on Mar 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
bdCLI tool and local bash scripts (e.g.,scripts/spawn-agents.sh) to manage feature tasks and agent processes. These actions are performed within the local project context to facilitate the development workflow.- [PROMPT_INJECTION]: The skill ingests user requirements and incorporates them into task descriptions and shell commands. This creates a surface for indirect prompt injection; however, the instruction to utilize a multi-step brainstorming and approval process before task creation serves as a functional mitigation strategy.- [SAFE]: Analysis of the skill instructions and code patterns revealed no evidence of remote code execution, obfuscation, persistence mechanisms, or unauthorized data exfiltration.
Audit Metadata